PROJECT SUMMARY Many of the up-and-coming therapeutic protocols in ophthalmology are technically difficult, near or beyond the limits of human ability, and are being attempted by only a few surgeons. Examples include subretinal injection of stem cells or gene therapies, which have the potential to restore sight but that cannot currently be delivered safely and repeatably, and the deep anterior lamellar karatoplasty (DALK) method for corneal transplantation, which eliminates the risk of transplant rejection but currently has a high failure rate. Surgeon hand tremor combined with patient movement due to breathing and snoring place a lower limit on achievable precision. The recent availability of optical coherence tomography (OCT), including new surgical microscopes with OCT embedded in the eyepieces, makes it easy to image the magnified retina and cornea, as well as the instrument relative to the tissue. However, positioning precision is the bottleneck. The investigators have developed a robotic manipulator that is more precise than all prior systems designed for robot-assisted eye surgery, and which is small and light enough to realistically be mounted to the head of the patient. The motivation for the prior support for the project was the potential benefit of head-mounting, with the hypothesis being that passive compensation of head movement (combined with the tremor canceling and scaling inherent to telemanipulation) will enable levels of precision that are currently unachievable by any means. However, to date, the benefit of head-mounting in robot-assisted ophthalmology has simply been assumed by the clinicians and engineers involved. Testing this hypothesis, and quantifying the benefit, is the critical remaining step that will be addressed in this project. This project comprises two specific aims, which are to quantify the positioning accuracy and precision of the head-mounted telemanipulated robotic system under OCT guidance, and compare to the results for manual abilities, for the applications of subretinal injection and DALK-method corneal transplantation. The specific aims will be addressed through human-subjects testing with experienced retinal and corneal-transplant surgeons. In order to conduct these experiments, the investigators will develop a robotic head that mimics the breathing and snoring of a patient, which can accommodate a pig eye in its eye socket. They will also develop a device to mount the robotic manipulator to the robotic head in a minimally invasive way, based on an existing device developed for surgical registration. The primary purpose of this exploratory/developmental project is to generate preliminary data to pursue R01 funding for two independent research projects involving robot-assisted subretinal injections of therapeutic agents and robot-assisted corneal transplantation, respectively.